SBTHCT: Segmentation of Brain Tissues using Hybrid Clustering Technique

Abstract

Because of the complex brain tumour structure, boring bodies and external factors like noise, brain magnet resonance imaging data have difficulty in influencing the tumour and oedema. Apart from the morphological operations, application of an effective hybrid clustering algorithm to segment brain tumors in this project is suggested to ease noise sensitivity and increase segmentation stability. Vienna adaptive filtration is particularly used for denoise and for the removal of non brain tissue morphology, thereby effectively reducing process sensitivity to noise. The most important contributions are: Second, the K-Man++ and Gaussian C-Fuzzy cluster refers to the algorithm of the segment images. This consolidation not only enhances the reliability and sensitivity of the algorithm. The tumor pictures removed are eventually processed after morphological procedure and median filtering in order to achieve correct brain tumor representation. The algorithm proposed was compared to other existing segmentation algorithms. The results show that the proposed algorithm is better accurately, sensitively, specifically and performance retrieval.